The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Recommends installing the @membranehq/cli tool, which is a trusted resource from the skill vendor (membranedev) for managing API integrations.
[COMMAND_EXECUTION]: Employs the membrane CLI for authentication and executing API actions, which are the intended and safe methods for this skill's functionality.
[PROMPT_INJECTION]: The skill processes transcription data from external audio files, representing an indirect prompt injection surface.
Ingestion points: create-transcript, get-transcript and search-words actions in SKILL.md.
Boundary markers: None explicitly mentioned in the skill text.
Capability inventory: Subprocess calls are limited to the membrane CLI which manages network and credential operations as described in SKILL.md.
Sanitization: Not described in the skill; however, the skill serves as a data retrieval tool and does not interpolate output into unsafe sinks.

assemblyai